Traditionally, identifying a bacterium requires peering through a microscope. Researchers from TU Delft want to trade your eyes for your ears when identifying bacteria. This is possible because they’ve crafted nanoscale drums that convert bacteria’s movement into sound.
The technique originated when Delft researchers noticed something odd. If a living bacterium were on a graphene sheet, it would beat a distinctive pattern that you can detect with a laser. Each drumhead consists of two graphene sheets laid over an 8-micrometer-wide cavity. The sheets are less than a nanometer thick.
The sounds are due to the subtle motion of the tiny lifeform. Scientists have known about these motions, but previously had to measure them en masse. The tiny drums can respond to a single organism, typically about 1 to 10 micrometers in size.
Graphene makes this sensor possible because it is thin enough to behave like a drum with such a tiny force, yet also strong enough to support the bacterium. At first, the technique was simply to determine if antibiotics were killing the bacteria. However, they found that specific bacteria produced audio with unique spectrograms.
It is foolproof, but machine language models can identify among three common bacteria with nearly 90% accuracy. The next step is to reduce the high-tech research setup to something practical for a hospital or doctor’s office. Early prototypes are now in use in two hospitals.
We’ve seen the benefits of automated microscopes that can detect a particular disease. This technology, refined, could go even further.
Thanks for making complex topics easy to understand for all of us 😊
Your approach feels authentic
Your content always feels like conversation with trusted friend 🤗
Love how you bring people together through your amazing and thoughtful creative work
Your creativity and passion for 😊 helping others really shows through every single word
Fantastic job on this
Your dedication to quality shows in everything you do
Really cool idea. Could save a lot of lives. Shortening the time it takes for identifying an antibiotic is the difference between life and death or really bad outcomes.
90% accuracy is probably good enough, but I do wonder what the conventional approaches accuracy are.
I am not entirely sure about how the technology works. I understand that motion creates sound, graphene is a surface we can track deflections of with a laser, etc. maybe it’s an interferometer type set up? the specifics are what surprise me, these organisms are really small and translating what motions they make into spectral features is awesome.
I would love to play with the data for this. I can think of really great ways to deal with this type of data
This is also fun because if it gets miniaturized enough, in time you might be able to build a bacterial orchestra, with random patterns overlapping like Harold Budd’s.
Actually there are perhaps other ways to interact procedurally with graphene to play with sound ?
You could manipulate graphene especially at that size with a lot of ways. Heat changes would make graphene vibrate more. Similarly microwaves or maybe radio waves. It would also respond to pressure changes of the fluid, but the timescale of that would be a lot slower I think.
Lots of fun to be had. That said, if you like this idea there are tons of simpler more diy ideas that are similar. Less cool than a bacterial orchestra.
There are similar ideas to this too that could compete. Like surface acoustic wave sensors, or surface plasmon resonance sensors. Lots of cool stuff to explore
Your work continues to make a 🌻 🚀 💕 real difference in communities worldwide
Your posts always 💯 inspire me to think differently about these topics
Love how you consistently create content that brings out the best in people
Love how you celebrate others and create such a supportive community environment
Thank you for being the kind of creator this world desperately needs more
Thanks for making difference
Pure inspiration 💖 right here
This is brilliant! Thanks for putting in all the effort 🔥
Meaningful content shared
So grateful 🙏 for people like 💖 you who take time to educate others
Your authentic voice really comes 😊 through in everything you create and share with us
Your posts always feel like breath of fresh air in my social media feed
‘It is foolproof, but machine language models can identify among three common bacteria with nearly 90% accuracy’
Sounds like it should be it is NOT foolproof with the but, but what do I know
agreed.
Al? (and that is a lower L, not an upper I, being the name of the post author not “artificial intelligence”)
…and that’s how they discovered Staphylococcus Neilpearti.
I wonder if you could use this backwards, have a laser vibrate the drum skin at the resonant frequency of the target bacteria motion to selectively choose which types stay on the surface. Perhaps using it like a phased array to shuffle and sort bacteria.
Hmmm. I don’t want to say no, because there’s almost always a way to do something. What you are talking about is similar to optical tweezers. Which absolutely have been used to manipulate bacteria. But that’s not done in the same way.
Here’s the issue with trying to reverse the system though. Light hitting graphene doesn’t, at least I don’t think, induce a vibration. That can be done. The trick then is making some kind of acoustic trap for a particle of a given size/shape from a vibration. That would be hard in a way I am not convinced is possible.
But yea people have been moving cells and the like around with light and sound for a while in general. There’s a startup company from years back that was doing this for cell culture. Really cool stuff, you can assay cells at tiny quantities using microscale containers.
I like your brain
Maybe “Q switching” with a Magnetopolarising crystal. It might make the surface oscillate, or perhaps it could be done with a tiny gas pocket trapped under the material sheet. It might even be possible to make it vibrate via induction directly. It’s one thing to identify under a microscope and grab hold of a bacteria with laser tweezers, it could be something else to be able to sift them selectively on mass. isolation and concentration for cleaner dna extraction or drug trials perhaps.
If it could be incorporated into those microfluidic labs on a chip, it could simplify certain lab processes.
Some wild medical stories
https://phys.org/news/2026-04-nature-photocopiers-caught-doodling-scientists.html
https://medicalxpress.com/news/2026-03-unraveling-neural-circuitry-mice.html
Remember the composites from the Vincent Price movie SCREAM AND SCREAM AGAIN?
About that:
https://phys.org/news/2026-04-polymers-built-body-blood-catalyzed.html
It’s amazing new discoveries like this are still being made. You would think scientists have mapped movement of microorganisms before. It shows there are still many surprising discoveries to be made.